A Real-time Cargo Damage Management System via a Sorting Array Triangulation (SAT)

Technique

1 Philip B. Alipour

2 Matteus Magnusson

3 Martin W. Olsson

4 Nooshin H. Ghasemi

Presentation by:

Problem Description

• 1. Cargo damage could happen anywhere• 2. Weak support system during transport• 3. Inspect and profile the damaged from undamaged• 4. Parameters are:

– cargo stack population n, – suggested containers stack for inspection S = (xyi) from

{xy1…, xyn}

– cargo status or detected Dxyz = Red or Grey or Green

– time to detect ti, total time to generate status TD, saved time Tsaved by SAT vs. time if used by other techniques, Tother.

• 5. Parametric adaptations: flocal = xyi coordinates for stack location, xyz for physical volume exposures (leakages) and damaged surfaces

Problem Design & Objectives

• Detect Defected or Damaged Cargos (ranging from low to severe)• Sort Damaged from Undamaged Cargos • Suggest Specific Containers from the Sorted List• Manage Cargos by human and non-human agents:

– Detectors mounted onto the crane facility in port;– Detectors stationed on the vessel after cargo load

• Finalize results as a report for a Final Managerial Decision, saying:– Don't get wined up with the details, rather matches of the results in the

database, inputs, outputs and updates coming from the algorithm. – Correct input gives correct output estimates probabilistically, either in

form of more estimates otherwise precisions.

Solution Method for DSS

• Our simulation method integrates regular and quasi-Monte Carlo methods for computing each triangulation phase:1. We Define a domain of possible inputs on i as (x1y1, …, xnyn)

2. Generate inputs randomly from a probability distribution over the domain (2D Array) giving xi * yi.

Random Function f(n) = Rnd(i)

We inspect cargo i for one triangle side (α),

Two more i’s for the remaining triangle sides (β and γ)

3. We generate an estimate between the detected 3, using distances to the remaining adjacent stacks in a database

4. We aggregate and update the results from one SAT phase to another on the whole population.

Solution Method for DSS

• The (red, grey, green) status ratio is then computed out of the whole stack population to generate the final status profile on cargos

• Optimization is mainly on the performance time relative to DB information updates and detection parameters. Here is an example from the main plan

Saved Time Effort:

Tsaved = Tother - TD

113 s

602.66 s

STATUS

• Current project status and problems:– Program results on the hybrid simulation– Profile generation and phase-to-phase triangulation

estimates on cargo status– Further analysis to our conducted SWOT analysis, like

sensitivity analysis on the time and spatial factors (performance)• This results in assessing and tackling Uncertainties in the

project• This further results in Verification and Validation of our

incorporated technique.

Q & A